The Cannabis Trademark Race: What the 2018 Farm Bill Tells Us About Schedule III

With cannabis rescheduling to Schedule III on the horizon, most of the industry conversation has focused on the 280E tax burden. That is the loudest pain point, but it is not the biggest long-term strategic problem for brand-builders. The bigger problem is cannabis trademark priority. A well-established cannabis brand operating since 2015 could very well find itself losing federal priority to a competitor who registered the same name the day the new rule takes effect.

This article is general information for brand owners and practitioners, not legal advice. I run GleanMark, a trademark intelligence platform built on top of USPTO's full database, and we have spent the last year looking at what actually happened to cannabis-adjacent filings after the 2018 Farm Bill. The data tells a clear story, and the same pattern is about to repeat.

What USPTO Actually Did With the 2018 Farm Bill

The 2018 Farm Bill removed "hemp" (cannabis with no more than 0.3% delta-9 THC on a dry-weight basis) from the Controlled Substances Act on December 20, 2018. But FDCA constraints stayed in place. CBD in foods and dietary supplements remained unlawful under FDA's position, and unapproved drug claims stayed barred. USPTO applies both the CSA and the FDCA when examining cannabis and CBD marks — this is spelled out in USPTO Examination Guide 1-19 (May 2, 2019) and TMEP §907.

You might expect that brands with years of state-legal CBD sales would get credit for their prior use. They did not. USPTO's position was firm: for goods that were previously unlawful, priority dates started on the Farm Bill's effective date. Nothing before December 20, 2018 counted for federal priority purposes for those newly lawful goods. For already-lawful adjacent goods and services (apparel, educational services, and the like), prior use continued to count as it always had.

This is not abstract. We looked at every trademark filing in GleanMark's database containing "cannabis," "marijuana," "CBD," "THC," or "hemp" filed between 2010 and 2025 (the 2025 cohort is preliminary — many of those applications are still in examination). Here is what the data shows:

Filing Year	Applications	Live Today	Pct Dead
2017	1,015	196	80.7%
2018	1,738	436	74.9%
2019 (Farm Bill year)	3,153	826	73.8%
2020	1,801	645	64.2%
2021	1,544	574	62.8%
2022	1,034	448	56.7%
2023	850	487	42.7%
2024	780	545	30.1%
2025	754	654	13.3% (preliminary)

Two things jump out. First, 2019 was the largest cannabis-adjacent filing year on record: 3,153 applications, an 81% spike over 2018. Filers rushed the gates the moment federal lawfulness became plausible. Second, the death rate stayed above 60% for years, because filers kept running into the same structural problems. It took the market several years of trial and error to figure out what actually gets registered.

When the Schedule III final rule takes effect, a similar pattern will likely unfold. For cannabis goods that were not federally lawful before the rule, pre-rule use will not establish federal priority. A day-one filer may obtain earlier federal priority for those newly lawful goods than a long-time state operator who delays — subject to any earlier federal or common-law rights those operators already hold in related lawful goods and services.

The Schedule III / FDA / DEA Wrinkle

Rescheduling from Schedule I to Schedule III does not magically unlock federal trademark registration for core cannabis goods. Here is what Schedule III actually does and does not do:

What rescheduling does:

• Permits controlled distribution among DEA-registered parties
• Makes most 280E tax deductions available (the tax angle everyone is celebrating)
• Opens research pathways
• Reduces regulatory penalties for certain categories of possession

What rescheduling does not do:

• Authorize consumer-facing interstate commerce in cannabis drug products without FDA approval
• Waive the need for DEA registration for continued distribution
• Make state-legal adult-use retail federally lawful — adult-use retail remains unlawful absent separate CSA authorization
• Remove cannabis products from the FDCA framework

USPTO's position, reinforced across the Farm Bill and prior guidance, is that a good or service must be lawful under federal law to register a mark for it (TMEP §907). Schedule III substances in consumer commerce generally require FDA approval. This means:

• Pharmaceutical cannabis products without FDA approval will still face "unlawful use" refusals
• Adult-use retail cannabis remains federally unlawful unless separately legalized
• Ancillary medical services that do not involve the manufacture, distribution, or dispensing of controlled substances (physician counseling, patient education, telehealth information) may be registrable; retail or dispensary services remain barred
• Cannabis-adjacent goods (apparel, paraphernalia-where-careful, education, software, media) remain the most reliable targets

The practical implication: if your brand sells flower or edibles, you will not walk into a 1A registration on day one of Schedule III for those core goods. You will still be in workaround territory for the substance itself until a compliant federal pathway exists.

The ITU Trap

A reasonable-sounding strategy is to file 1B intent-to-use applications now for cannabis products you expect to sell post-rescheduling, preserve priority, then convert to 1A once federal lawfulness is established. This strategy has a structural problem.

USPTO has consistently rejected 1B applications where the applicant's intended use is per se unlawful at the time of filing, on bona-fide-intent grounds. The TTAB has addressed this directly. In In re Brown, 119 USPQ2d 1350 (TTAB 2016), the Board affirmed refusal of "retail marijuana store services" on the ground that the services were unlawful under the CSA. In re JJ206, LLC, 120 USPQ2d 1568 (TTAB 2016), and In re PharmaCann LLC, 123 USPQ2d 1122 (TTAB 2017), applied similar reasoning — the lawful-use doctrine bars registration for both 1(a) use-based and 1(b) intent-to-use applications where the underlying goods or services are federally unlawful. USPTO Examination Guide 1-19 confirms this: ITUs covering unlawful goods will be refused.

Filing an ITU today for "retail sale of medical cannabis" or "distribution of marijuana products" is almost certainly a losing ticket. You will burn filing fees and, worse, telegraph your brand strategy to competitors who can see the published application.

The ITU strategy works only for goods and services that are genuinely lawful at the time of filing. Which brings us to the playbook.

The Practical Prep Playbook

If you are a brand holder, here are the steps to run before Schedule III takes effect. Knockout search your target name first — if someone else already owns the brand in your space, none of the rest matters.

1. File now for every lawful adjacent good and service you can defensibly claim.

Based on our filing data, the classes that survive most reliably for cannabis-adjacent brands are:

Nice Class	What It Covers	Post-Farm-Bill Survival Rate
036	Financial services (e.g., cannabis industry banking, investment)	73%
042	SaaS, technology, scientific services	67%
041	Educational services, training, content	66%
044	Medical services (ancillary, non-dispensing)	63%
009	Downloadable software, apps	63%
035	Retail services — lawful retail (apparel, general merchandise), business consulting. Not "retail cannabis store services," which remain refused.	57%
025	Clothing, apparel	51%
003	Cosmetics, skin care	41%
034	Tobacco, smokers' articles, paraphernalia. Avoid identifications that imply cannabis use; 21 U.S.C. §863 can bar marks for drug paraphernalia.	36%
005	Pharmaceutical preparations	19%

The pattern is sharp. Class 005 (pharmaceuticals) has the lowest survival rate because cannabis products in that class are most likely to trigger "unlawful use" refusals. The adjacent-services classes (035, 041, 042, 009) survive because the services themselves can be lawful even while the underlying substance is not. If your brand portfolio does not yet include apparel, retail-of-lawful-goods, and education-adjacent filings, you are leaving priority on the table.

2. Build documentary records of state-legal use — for the right reasons.

For every state your brand operates in, keep rigorous records of first use dates, sales volumes, customer counts, marketing, advertising, and geographic footprint.

For cannabis goods that are not federally lawful yet, these records will not support federal priority. What they will support is common-law rights, state-registration claims, and any future argument that your brand has priority in specific geographic markets. For already-lawful goods and services (apparel, education, software), these records do support federal priority as they always have. This is defensive infrastructure for both worlds.

3. Have filings drafted and ready to file as soon as a lawful pathway exists — but file the right basis for each good.

The day-one race matters, but the filing basis matters more. For already-lawful adjacent goods and services (apparel, software, education, ancillary medical, etc.), file 1(a) now or as soon as you have use in commerce. There is no reason to wait. For core cannabis goods whose lawful sale depends on DEA and FDA compliance, expect to file 1(b) intent-to-use only once there is a lawful federal pathway — and understand that until that pathway exists, USPTO is likely to continue refusing those filings on bona-fide-intent grounds. Your attorney should have the draft applications ready so that the moment the regulatory picture changes, you are filing the same day.

4. Layer state-level trademark registrations.

State trademark registrations have been underused by cannabis brands. With federal registration locked out for core goods, state registrations become your primary offensive tool. They are cheap, fast, and provide a clear paper trail of your brand claims. Priorities here: the states where you actually sell, plus any state with a developed TTAB-equivalent process that could support enforcement actions.

5. Monitor the competitive landscape aggressively.

The race means new filers will be coming in. You need to know when competitors file applications that might conflict with your brand so you can act within the opposition window. This is what trademark monitoring is for. For cannabis brands specifically, the monitoring volume is going to spike dramatically in the months after rescheduling. GleanMark handles this kind of ongoing watch across the full USPTO database of over 14 million records. Knowing what competitors are filing, the same business day they file it, is how you stay ahead of the opposition window rather than chasing it.

What USPTO Can and Cannot Do Administratively

A fair question from practitioners: should USPTO offer some kind of protection for brands with prior state-legal use?

The honest answer is that USPTO cannot alter the lawful-use requirement by guidance alone. The "use in commerce" standard is statutory, and the lawful-use doctrine is entrenched in TTAB and court precedent. Meaningful change would likely require either congressional action or a shift in controlling court precedent. To date neither the SAFE Banking Act nor the 2018 Farm Bill included cannabis trademark reform, and the TTAB has repeatedly reaffirmed the doctrine.

What USPTO could do administratively is publish clearer guidance post-rescheduling on which goods are in-scope and which are not. The current guidance on cannabis-adjacent goods is scattered across USPTO Examination Guide 1-19 and multiple TMEP sections, and examiner practice varies. Post-Schedule-III, a fresh examination guide would reduce wasted filing volume. That is incremental; it does not solve the priority-reset problem.

Realistically, established brands will need to rely on common-law rights, dominant domain presence, and state-level registrations as their first-mover protection, with federal filings queued for the day rescheduling becomes operative. For an overview of how the examination process works once those filings go in, the process is the same as any other trademark application — what changes is the opening date of the window.

The Bottom Line

Cannabis rescheduling is a priority event for trademark holders. The 2018 Farm Bill is a real-world case study of what happens when federal lawfulness expands for a category of goods: filings spike, priority resets for the newly lawful subset, and long-time state-legal operators find themselves competing with day-one filers for their own brand names on those specific goods.

The brands that navigate the post-Schedule-III window well will be the ones that:

• File 1(a) now for every lawful adjacent class (36, 42, 41, 44, 9, 35, 25 lead on survival)
• Avoid ITU filings for goods that are unlawful today — TTAB will refuse them
• Keep documentary records that support common-law and state-priority claims
• Have draft applications ready for the moment a federal pathway opens for core cannabis goods
• Monitor the filing landscape aggressively as the rule takes effect

For a foundational overview of how the trademark system works before getting into strategy, trademark basics is a good place to start. If you are already operating and trying to think through multi-class coverage, the 45-class coverage analysis has useful data on how much protection to pursue at different brand stages.

This article is general information, not legal advice. The lawful-use doctrine, FDA and DEA frameworks, and USPTO examination practice all have case-specific nuances. Work with qualified trademark counsel to tailor your filing bases, identifications, and timing to your actual brand and the current regulatory landscape. The planning window is now.